Stool-derived Extracellular Vesicles Increase Inflammasome Signaling and Regulate the Gut-Brain Axis after Stroke in Alzheimer’s Disease Transgenic Mice

Abstract

Patients with Alzheimer’s disease (AD) suffering from post-stroke gut dysfunction present with worsened neurological outcomes. This study investigated the role of stool-derived extracellular vesicle (EV)-mediated inflammasome signaling in the gut-brain axis following photothrombotic stroke (PTS) in aged 3xTg- AD and wildtype (WT) mice. Western Blot and immunohistochemical analyses evaluated inflammasome signaling proteins, Gasdermin D (GSDMD), and Aβ in intestinal and cortical tissues. Gut permeability was measured using a FITC-dextran assay 3 days post PTS. Adoptive transfer experiments assessed the impact of stool-derived EVs from PTS mice on inflammasome signaling in recipient naïve 3xTg and WT mice. At 3 days, 3xTg-PTS mice demonstrated significantly impaired sensorimotor Rotarod performance compared to WT-PTS mice. Both WT and 3xTg PTS mice had deficits compared to 3xTg and WT sham mice using the Open Field or Novel Object Recognition tests. Compared to WT- PTS mice, 3xTg-PTS mice had disrupted gut morphology at 1-month post-PTS, as well as increased gut permeability at 72 hours. Immunohistochemical analysis also revealed activated microglial morphology and presence of GSDMD and Aβ in the brain and intestines post-PTS in 3xTg and WT mice. Adoptive transfer of stool-derived EVs from PTS mice to WT mice induced elevated levels of inflammasome signaling proteins in recipient cerebral cortices. These findings indicate an important role of stool-derived EV inflammasome signaling and pyroptosis in disruption of the bidirectional gut-brain axis after stroke leading to exacerbation of AD pathology in aged WT and 3xTg mice.

Graphical Abstract.

Schematic showing that after stroke in preexisting AD, serum-derived EVs containing inflammasome proteins are released into the circulation and taken up into the gut. In turn, increased gut permeability and pyroptosis leads to the release of stool-derived EVs from the gut containing inflammasome proteins, are released into the circulation and taken up by the brain causing increased neuroinflammation, pyroptosis, and worsened neurological function.

Introduction

Stroke is the second leading cause of death worldwide with an annual mortality rate of about 5.5 million with 50% of stroke survivors being chronically disabled (Donkor, 2018). Alzheimer’s disease (AD) is ranked as the 7^th most common cause of death in the United States (Vijayan & Reddy, 2016). Risk factors that contribute to the development of AD may include a variety of factors including genetic predisposition, age, cerebrovascular disease, diabetes, and obesity (Silva et al., 2019). Stroke and AD often co-exist, but the effect on stroke outcomes, AD genetic factors and systemic organ complications are multifactorial and understudied (Waziry et al., 2020). Clinical studies have shown that AD patients are more vulnerable to cerebrovascular disease compared to non-AD control patients (Subic et al., 2017). Stroke has been shown to contribute to AD neuropathology such as accumulation of abnormal proteins including amyloid-beta (Aβ). Moreover, an increase in Aβ accumulation has been associated with impairment of vascular physiology (Charidimou et al., 2017).

It is well known that aging is one of the most common risk factors for both AD and stroke (Wang et al., 2021). AD is characterized by dementia with age-related decline, and stroke frequently presents itself in older populations. The 3xTg mouse model has been well established for chronic studies investigating the progression of AD pathology (Javonillo et al., 2021). At 10 to 12 months of age, 3xTg mice already express Aβ plaques and start to develop tau tangles (Belfiore et al., 2019). Both AD and stroke patients experience systemic organ complications, such as gastrointestinal (GI) complications, and these complications can progress with age (Dumic et al., 2019). Examples of GI disorders such as irritable bowel syndrome (IBS) and constipation, are seen in both AD and stroke patients (Arya & Hu, 2018; Li et al., 2017; Nakase et al., 2022; Wang et al., 2001). After stroke, 50% of patients experience GI problems including dysphagia, constipation, fecal incontinence, and GI bleeding (Camara-Lemarroy et al., 2014). Stroke patients can also have an impaired nutritional status, which is associated with reduced functional outcome (Schaller et al., 2006). Interestingly, stroke patients with GI complications are more likely to suffer worse neurological outcomes, suggesting a bidirectional nature between the gut and the brain (Daneman & Rescigno, 2009).

The interaction between the gut and neurological issues such as stroke and AD are complex and has been shown to be bidirectional (Suganya & Koo, 2020). Alterations to the gut microbiome have been reported in AD patients, including pro-inflammatory associated bacteria (Haran et al., 2019; Zhuang et al., 2018), and a growing body of research suggests that therapeutically targeting gut microbial composition may be beneficial in mitigating the effects of stroke as well as AD (Murray et al., 2022).

Our previous studies have reported that inflammasome-induced cell death plays a significant role in disruption of the gut-brain axis after stroke (Kerr et al., 2022). Moreover, the release of Extracellular Vesicles (EVs) after stroke (N. Kerr et al., 2018) and traumatic brain injury (de Rivero Vaccari et al., 2016) carry a cargo of inflammasome proteins that induce a systemic inflammatory response in peripheral tissues (Cyr et al., 2024; Keane et al., 2023; Kerr et al., 2020; Kerr et al., 2019; Kerr et al., 2021). The inflammasome is a multi-protein complex involved in the activation of caspase-1 leading to the maturation of IL-1β and IL-18 (Li et al., 2021). The activation of caspase-1 leads to a regulated form of programmed cell death, through cleavage of Gasdermin-D (GSDMD), known as pyroptosis (Yu et al., 2021). Pyroptosis is characterized by a pore formation in the plasma membrane leading to the rupture of the membrane. Furthermore, inflammasome activation is positively correlated with microglial activation in the brains of AD transgenic mice, (Shukla et al., 2021) and inflammasome-mediated pyroptosis has been implicated in the progression of neuroinflammation after stroke (Abulafia et al., 2009; Chen et al., 2020). In addition, inflammasome signaling and pyroptosis have been found to have a mechanistic role in chronic gut disorders, such as irritable bowel syndrome, inflammatory bowel disease and gut dysbiosis, all of which are found as GI complications in stroke and AD patients (Heston et al., 2023; Song et al., 2023).

The purpose of this study was to test the hypothesis that stroke in aging and AD induces the release of stool-derived EVs that disrupt the gut-brain axis and induce pyroptosis in the brain after injury. Previous studies in our laboratory have examined the transport of EVs as a signaling mechanism between the brain and distal organ damage after traumatic brain injury and stroke (N. A. Kerr et al., 2018; Kerr et al., 2022). We have also reported that inflammasome-related proteins are elevated in serum-derived EVs of stroke patients (N. Kerr et al., 2018). For this study we used male and female mice that were 10–12 months of age due to the importance of aging as a factor for developing stroke and the role that the inflammasome plays on inflammaging (Cyr & de Rivero Vaccari, 2023b; Cyr et al., 2022; Mejias et al., 2018; Raval et al., 2019).

Materials and methods

Animals

All experimental procedures for this study were approved by the University of Miami's Animal Care and Use Committee and were conducted in compliance with the ARRIVE guidelines and those established by the National Institute of Health Guide for the Care and Use of Laboratory Animals (Percie du Sert et al., 2020). Animals were housed in a temperature-controlled room (22°C) with a 12-hour light/dark cycle. They had at least 7 days of acclimation time before undergoing any experimentation. All animals had access to food and water ad libitum. For this study, 10–12-month-old 3xTg and WT-B6129 mice were used. In accordance with NIH guidelines, male and female mice were used for this study. In 3xTg mice, because females lose weight with aging and males tended to gain weight (Robison et al., 2023), including both sexes in this study produced a range in weight between 25–40 g. The 3xTg mouse model is well established to study aging and the progression of AD since they develop AD pathology after 6 months of age (Javonillo et al., 2021). Each study consisted of the following four groups that were randomly assigned: WT-Sham, WT-PTS, 3xTg-PTS, and 3xTg-Sham. Each experiment with time points and groups is described in Figure 1. To ensure rigorous, accurate and unbiased results, power analysis was carried out based on prior studies conducted in the photothrombotic stroke (PTS) model to determine the appropriate sample size for the study (Kerr et al., 2022), hence ensuring that the study was adequately powered to produce meaningful and reliable results.

Figure 1. Experimental timeline and description of methods and animals used for each study.

The following groups were used for each study WT-sham, WT-PTS, 3xTg-sham, 3xTg-PTS. In study 1, mice were sacrificed 24 hours after PTS for protein expression analysis. In study 2, mice were sacrificed at 3 days after PTS for gut permeability assessment. In study 3, mice were sacrificed 1 week after PTS for behavioral and histological assessment.

Photothrombotic Stroke (PTS)

Male and female 3xTg-transgenic AD (10–12 months of age and between 25–40 grams) and corresponding wild-type mice (B6129) were anesthetized using ketamine (100 mg/kg) and xylazine (10 mg/kg) and placed in a stereotaxic frame to immobilize them. Throughout the surgery, the head and rectal temperature were maintained at 37± 0.5°C. A small surgical midline incision was made in the skin to expose the skull. Next, Rose Bengal, a photochemical dye, was slowly injected via the tail vein at a concentration of 15mg/ml, with a dosage of 10ul/g of animal weight. Using a 532nm YAG laser (Laserglow) with a 3mm aperture, a photothrombotic cerebral infarction was created by positioning the laser beam above the right cerebral cortex, approximately 2mm lateral from Bregma (Kerr et al., 2022). Next, the skull was irradiated for 9 minutes at 25 mW to induce a closed head PTS (22). Post-irradiation, mice were allowed to regain consciousness and buprenorphine (0.1mg/kg) was administered for pain. Sham-operated animals underwent identical surgical procedures and Rose Bengal injection, but they were not subjected to irradiation. After the surgery, all animals were allowed to survive for varying periods.

Behavioral Testing

To test locomotor and anxiety-like behavior after stroke and AD, mice were subjected to the Open Field test 72 hours after PTS. Prior to testing, mice were acclimated in the testing room for 20 min. After the acclimation period mice were then pre-handled and tested in an open field task for 10-minute intervals in a plexiglass box (43.2-cm width, 43.2-cm length, 30.5-cm height) with transparent walls and a white floor (Johnson et al., 2023; Kerr et al., 2022). Mouse behavior was recorded using an Ultra 720+ Resolution DSP, True Day/Night Color Camera (Everfocus, Duarte, CA, USA) and analyzed using EthoVision XT Version 10.0 software (Noldus, Asheville, NC, USA). The mouse’s movements, including speed and total distance traveled, were recorded.

To measure sensorimotor function, the Rotarod test was assessed on days 3-, 7- and 14-days post-stroke (Kerr et al., 2022). Following established parameters, mice were placed on a horizontal rotarod cylinder. The duration that mice stayed on the rotarod was measured in seconds. This measurement is indicative of their balance, coordination, and physical condition. The rotarod apparatus was set in the accelerated mode. In this setting the speed increased from 4–40 rpm over the course of 600 sec and was repeated 4 times for each animal, with resting intervals of 2 minutes between each trial. To avoid potential time-of-day effects, rotarod testing was conducted at the same time of the day. Pretraining was conducted for 3 consecutive days before undergoing PTS. The average duration (in seconds) on the machine was calculated from the 3 different rotarod measurements 24 hours prior to surgery.

Fourteen days after PTS, mice underwent the novel object recognition (NOR) task over 3 days. Each testing day, mice were habituated in the room to an open field arena for 10 minutes. On the second day, mice were placed in the same environment with two identical objects that were placed equidistantly from each other and were allowed to explore for 5 min. On the third day, mice explored the same environment for 10 minutes, with one of the familiar objects replaced by a novel object. Results were based on the amount of time mice spent exploring the novel object versus the familiar object. Analysis was performed using a video tracking system EthoVision (Noldus, Leesburg, VA) and showed that a −0.5 index indicates a preference for one object relative to the other after stroke based on the following equation: (Novel-Old)/(Novel + Old)= Positive= NOR, Negative= no NOR, 0= equal recognition. The preferential exploration of a new object is identified as the discrimination index and is a measurement of recognition memory sensitivity (Sivakumaran et al., 2018).

Histology and Immunohistochemistry

After behavioral testing, approximately 1-month post-stroke, mice were sacrificed for histological and immunohistochemical analysis. Brain tissue was collected and then placed in 10% formalin. The specimens were stored at room temperature for 48 hours before being transferred to 70% ethanol for 24 hours. Subsequently, the specimens were processed for paraffin embedding. Coronal tissue sections were acquired at 10 μm thickness on a microtome Leica HM 2125 at 150 μm intervals. One series of sections was stained with hematoxylin-eosin (H&E). Images were acquired at 4x using an Olympus BX51 microscope, and volume estimation was carried out by a blinded investigator using Neurolucida software (version 7.50.1, Micro Brightfield Inc., Williston, ME). Coronal brain sections were stained with H& E and imaged using brightfield microscopy with a 5× objective on an Axiophot 200 M microscope (Zeiss Microscopy, LLC, Thornwood, NY). The cortical hemispheres (left and right) were manually contoured by a blinded observer using Stereoinvestigator software (MBF Bioscience, Williston, VT) in an unbiased manner. Serial sections were analyzed beginning at Bregma +1.33 mm, with every other section sampled caudally through Bregma −2.91 mm. Cortical volume asymmetry was quantified using a standard asymmetry index, reported as a percentage difference, and calculated as:(Left Volume−Right Volume) / ((Left Volume+Right Volume)/2) X100. All four experimental groups were included in this volumetric analysis.

Intestinal tissue was collected and fixed from the same group of mice 1-month post-stroke as well. Following the abdominal incision and transection of the colon, the entire length of the small intestine and colon was freed and washed in phosphate-buffered saline (PBS) to remove intestinal contents. In order to fix the intestinal tissue, Bouin’s fixative (Thermo Fisher Scientific, Waltham, MA) was used, and the Swiss roll method was employed to maintain gut morphology (Bialkowska et al., 2016). Slowly, the colon and small intestines were rolled around a toothpick with the luminal side facing up. The roll was transferred to a cassette in 10% buffered formalin for 48 h at room temperature. Finally, the tissue was transferred to 70% ethanol, as mentioned above. Sections were paraffin embedded and sections for H&E and immunohistochemical analysis.

In separate sections, intestinal and brain sections were double stained with primary antibodies using rabbit anti-amyloid beta (Novus Biologicals, Centennial, CO #NBP1–78007), anti-Iba1 (Abcam, Waltham, MA #EPR16588), or Epcam1 (#GTX636998, GeneTex, Irvine, CA). Secondary antibodies, all conjugated to Alexa Fluor dyes (488 #A20185 and 594 #A20181) were purchased from Invitrogen. Briefly, sections were deparaffinized in Xylene then rehydrated prior to antigen retrieval. Sections were then rinsed in PBS, permeabilized with 1% Triton X-100 (Sigma-Aldrich, St. Louis, MO), rinsed with PBS, then blocked with a solution of 5% normal goat serum, 2% BSA and 0.3% Triton X-100 in PBS for four hours, Sections were then incubated overnight with the specified primary antibodies diluted at required concentration in blocking buffer. For negative controls, sections were incubated in blocking buffer without a primary antibody. Next, sections were rinsed with PBS and incubated with secondary antibodies in blocking buffer for two hours at room temperature in the dark, then rinsed with PBS before mounting with Vectashield Mounting medium with DAPI (Vector Laboratories, Newark, CA). Confocal microscopy (Dragonfly 202, Andor Technology Ltd. Concord, MA) was used for imaging.

Gut Permeability Assay

In another group of mice (WT-Sham, WT-PTS, 3xTg-Sham, 3xTg-PTS) intestinal permeability was assessed three days post-PTS using oral gavage administration of FITC-dextran 4000 (Sigma-Aldirch, St. Louis, MO) (Woting & Blaut, 2018). On the day of the gavage, food and water were withdrawn for 4 hours, and then 150 ul of 80 mg/ml FITC-dextran (4 kDa) was administered through gavage feeding. Next, serum was collected after 4 hours and FITC-dextran measurements were performed in triplicates by fluorometry (excitation, 490 nm; emission, 530 nm). Serial dilutions of FITC-dextran in PBS were used to calculate a standard curve.

Western Blot Analysis

For biochemical analysis, in a separate group of mice, (WT-Sham, WT-PTS, 3xTg-Sham, 3xTg-PTS) (Fig 1), intestinal and cortical tissue samples were collected 1 day post PTS. Tissue samples were snap-frozen in liquid nitrogen immediately after removal. Two-millimeter sections of small intestine and right cortical tissue were homogenized in RIPA buffer containing a protease and phosphatase inhibitor cocktail (Sigma-Aldirch, St Louis, MO). Samples were then resolved by immunoblotting as described in (Cyr & de Rivero Vaccari, 2023a). Briefly, in 4–20% Tris-TGX Criterion gels (Bio-Rad, Hercules, CA) using antibodies to Caspase-1 (Novus Biologicals), ASC (Santa Cruz, Dallas, TX), IL-1β (Cell Signaling), and GSDMD (Santa Cruz, Danvers, MA). Image Lab software was used for quantification and all data were normalized to β-actin.

EV isolation

EVs were isolated from stool of PTS and sham mice 24 hours after injury (same group as Western Blot Analysis) using the Total Exosome Isolation (TEI) solution for other bodily fluids (Invitrogen, Carlsbad, CA). Based on previous studies (Northrop-Albrecht et al., 2022), stool samples were collected 24 hours post-PTS and frozen at −80 degrees Celsius. Before isolation, stool was thawed at room temperature. Once thawed 2 stool pellets per sample were diluted in 100 uL PBS. To ensure that stool was liquified a hand homogenizer was used. 100 μL of each sample was centrifuged at 2000 ×g for 30 minutes. The supernatant was then incubated with 20 μL of TEI for 30 min at room temperature. After incubation the samples underwent centrifugation at 10,000 × g for 10 min. The pellet was resuspended in 100 μL of PBS. EVs concentration and size was determined using Nanoparticle Tracking Analysis (N. A. Kerr et al., 2018).

Electron Microscopy of stool-derived EVs

Stool EVs were isolated as described above. To evaluate ultrastructural characteristics, samples were loaded onto a carbon copper grid for 30 minutes, rinsed in phosphate buffer, then double-distilled water, fixed with 2% glutaraldehyde, and stained with a 2% aqueous uranyl acetate solution. Grids were kept overnight protected from light and viewed at 80 kV in a JEM-1400 transmission electron microscope (JEOL, Peabody, MA) and images captured with an AMT BioSprint 12 digital camera.

Adoptive transfer of EVs

Stool-derived EVs from 3xTg-PTS and WT-PTS were injected into 3xTg-Naïve and WT-Naïve mice, respectively. Stool-EVs were isolated from mice sacrificed at 24 hours after PTS. EVs were injected into anesthetized (ketamine/xylazine) naïve mice (n=5) through the tail vein at a dose of 1.0 × 10¹⁰ particles/g/body weight (N. A. Kerr et al., 2018; Kerr et al., 2022; Wiklander et al., 2015). Particle count was measured by Nanosight Analysis and samples were diluted as described above. Mice were sacrificed 24 hours after adoptive transfer and right cortical tissue was collected for western blot analysis.

Statistical analysis

Data were analyzed using a student’s T-test for two groups and a one-way ANOVA followed by student Newman- Keuls test (GraphPad Prism version 7.0) for two or more groups. D’Agostino-Pearson test was used to test for normality. Data are expressed as mean +/− SEM. P values of significance used were * p<0.05. Statistical analyses were performed using Prism 9.0 (GraphPad Software, Inc., La Jolla, CA, USA). Data were normally distributed, using a D’Agostino-Pearson test for normality. A sample size of 5–6 was used for all studies except behavior studies (8–10 mice per group) based on power analysis (using G* power analysis, with an effect size F = 0.85, α set at 0.05) and historical data (N. A. Kerr et al., 2018). Brain and intestinal sections were masked with opaque tape and analyzed by an individual who was blinded to the groups. To increase scientific rigor, randomization to the study groups for behavioral analysis was performed by a research assistant who was not involved in behavioral analysis. In addition, study investigators were unaware of the randomization until study completion. For this initial study group numbers were not large enough to analyze sex differences.

Results

Infarct size and mortality in aged WT and AD mice 1-month post-stroke

Our previously published studies reported that the PTS model produces a well demarcated cortical infarct in young mice 3 days after injury (Kerr et al., 2022). This closed skull model is advantageous for several reasons including the contributions of endothelial damage, local platelet activation and occlusive vascular thrombosis leading to severe reductions in focal cerebral blood flow in the pathogenesis of infarct formation. In this study, we induced PTS in aged 3xTg and WT mice. Here we show the size of the cortical infarct at 1-month post-PTS in both WT and 3xTg aged mice (Figures 2A and B). Histological analysis revealed slightly larger cortical infarcts in 3xTg-PTS mice compared to WT-PTS mice at 1 month (Figures 2A and B). In our previous studies using young adult WT mice, we demonstrated that the PTS model produces a general bowl-shaped cortical lesion without overt damage to the hippocampus. However, in our 3xTg-aged mouse model, subcortical damage in 3xTg mice occasionally involved the underlying hippocampus at multiple bregma levels (Figures 2A and B). We also found a significant increase in cortical volume asymmetry index, a measurement in the degree of asymmetry between the right and left hemispheres of the brain, in 3xTg-PTS mice compared to WT-PTS mice (Figure 2C). Interestingly, when examining survival rates, we observed no significant difference between aged PTS 3xTg and WT mice. However, there was an increase in the percent survival rate of the WT sham mice when compared to the 3xTg sham animals (Figure 2D). Taken together, these results support an important role of aging in the progression of stroke pathology and survival after injury in both 3xTg and WT mice.

Figure 2. Infarct size is similar in 3xTg-PTS mice compared to WT-PTS mice and mortality is increased after PTS in Aged WT and 3xTg Mice.

A-B) Aged 3xTg mice present larger infarct size compared to WT mice. C) Percent survival in aged 3xTg PTS mice is not decreased compared to WT PTS mice. WT-PTS percent survival is decreased compared to WT-Sham.

PTS leads to motor and cognitive dysfunction in aged 3xTg and WT mice

It has been reported that the PTS model produces significant deficits in motor and cognitive function in young healthy rodents within the first two weeks after injury (Kerr et al., 2022). Therefore, we examined changes in motor and cognitive function after PTS in aged AD and WT mice; the same groups were used for histological analysis (Figure 3). We used the Rotarod test to measure sensorimotor deficits after PTS in 3xTg and WT mice and in sham 3xTg and WT mice (Figure 3A). At 3 days after PTS there was a decrease in the time on the rod in 3xTg-PTS mice compared to the WT-PTS mice. However, at 7- and 14-days post-PTS, there was no significant difference in the time on the rod for the 3xTg-PTS and WT-PTS (Figure 3A). Taken together, these findings indicate that under the current experimental conditions, AD transgenic mice have increased motor deficits at early postischemic periods after stroke compared to WT aged mice.

Figure 3. PTS in 3xTg and WT mice produces motor and cognitive deficits.

A) 3xTg-PTS mice show decreased motor function 3 and 14-days post PTS compared to WT-PTS, 3xTg and WT sham mice on the Rotarod test. B) WT-PTS and 3xTg-PTSmice demonstrated learning and memory deficits as well as decreased exploratory behavior compared to sham mice using NOR test. C-D) 3xTg -PTS mice show significantly decreased distance moved, velocity on open field test. E-F) 3xTg and WT PTS mice time spent significantly more time in the border and less time in the center compared to 3xTg and WT sham mice. (n= 6–8 per group), *p<0.05, **p<0.01, ***p<0.001, One-Way ANOVA.

Next in the same group of mice, we performed Open Field testing 3 days post stroke and Novel Object Recognition (NOR) at 14 days post stroke. The NOR test is used to measure spatial and nonspatial memory as well as temporal order memory (Antunes & Biala, 2012). We found that there was no significant difference in discrimination index between aged WT and 3xTg PTS mice. However, both groups performed significantly worse compared to 3xTg and WT Sham mice (Figure 3B). Furthermore, the open field, which is used to measure exploration and anxiety, demonstrated that 3xTg-PTS mice moved significantly less in total distance and had decreased velocity compared to WT-PTS mice (Figure 3C). Both 3xTg-PTS and WT mice spent significantly more time in the border and less time in the center compared to 3xTg-sham and WT-sham mice (Figures 3E and 3F). Taken together these data suggest that both aged WT and 3xTg-PTS mice suffer from more anxiety-like behaviors compared to aged sham mice. These findings are consistent with clinical data showing that AD patients who suffer from dementia are also more likely to have worsened cognitive decline after stroke (Al-Qazzaz et al., 2014).

Inflammasome protein expression is increased in cerebral cortical tissue of 3xTg and WT-PTS mice and 3xTg sham mice

To determine the protein levels of inflammasome activation after PTS, we performed western blot analysis of ipsilateral (right) cortical tissue lysates of mice (Figure 4A). Our results show a significant increase in inflammasome proteins cleaved caspase-1 (20 kDA) (Figure 4B), cleaved ASC (25 kDA) (Figure 4C), and mature IL-1β (17kDA) (Figure 4D) in 3xTg-PTS mice compared to WT-PTS mice and 3xTg-Sham mice. Moreover, there was a significant increase in mature IL-1β in WT-PTS compared to 3xTg-Sham mice at 24 hours post injury (Figure 4D). There was a significant increase in cleaved GSDMD (25 kDA) in 3xTg-PTS mice compared to 3xTg-Sham mice. Additionally, 3xTg-Sham mice and WT-PTS mice exhibited similar levels of inflammasome protein expression. These findings suggest that a stroke in aged WT mice can lead to an increase in inflammatory markers that are similar to AD transgenic mice.

Figure 4. PTS induces inflammasome activation and GSDMD cleavage at 24 hours post-PTS in 3xTg and WT mice and 3xTg sham mice in cortical tissue.

A) Representative images of Western Blot analysis showing protein levels of inflammasome proteins in WT and 3xTg mice with and without PTS. Western blot analysis showing significant increase of cleaved caspase-1 (B), cleaved ASC (C) in cortical tissue 24 hours after PTS in aged 3xTg and WT mice and mature IL-1β in aged 3xTg- PTS compared to WT-PTS and 3xTg sham mice (D). E) Cleaved GSDMD is significantly increased in cortical tissue of aged 3xTg mice compared to 3xTg sham mice 24 hours after PTS. (n= 5–6 per group), *p<0.05, **p<0.01, ***p<0.001, One-Way ANOVA.

In addition, we performed immunohistochemical staining for microglia (Iba1) and Aβ on brain sections of aged 3xTg and WT mice at 1 month post stroke (Figure 5). Our findings indicated an increase of Aβ immunostaining in WT-PTS mice (Figure 5C) compared to WT-Sham (Figure 5A). Furthermore, Iba1 staining provided consistent evidence of microglial ameboid morphology in all groups, which may be due to aging (Cyr & de Rivero Vaccari, 2021). Interestingly, in 3xTg-PTS mice, there was morphological evidence of amoeboid microglia compared to the 3xTg-Sham and WT-PTS mice at 24 hours after PTS. Lastly, to support IHC results, we performed Western Blot analysis in cortical tissue 24 hours after PTS. Our results demonstrate a significant increase in Aβ42 in 3xTg-PTS compared to 3xTg- Sham mice (Figures 5E and F). However, there was no significant difference in between 3xTg-PTS and WT-PTS mice (Figures 5E and F). These findings suggest that aging, even without a genetic predisposition can enhance AD pathology after stroke.

Figure 5. PTS leads to microglial activation and accumulation of Aβ 1 month after PTS.

Confocal microscopy images showing expression of Aβ (red) and microglia (Iba1, green) in 3xTg-PTS (D), 3xTg-sham (B), and WT-PTS (C) mice compared to WT-sham mice (A) 1 month after stroke. 60 × images. Scale bar = 10 μm.

Inflammasome protein expression is increased in intestinal tissue of 3xTg-PTS and WT-PTS mice and 3xTg sham mice

To examine the role of pyroptotic intestinal cell death after PTS and in AD, we induced cerebral ischemia to aged WT and 3xTg mice and measured inflammasome proteins and GSDMD expression at 24 hours after PTS (Figure 6A). Our results demonstrate a significant increase in cleaved form caspase-1 (Figure 6B), the initiating protein for pyroptosis, cleaved ASC (Figure 6D) and mature form of IL-1β (Figure 6C) in 3xTg-PTS compared to WT-PTS mice. However, there was no significant difference in any of these proteins between WT-PTS mice and 3xTg-Sham mice (Figure 6A to C). In addition, there was a significant increase in cleaved GSDMD expression in the intestinal tissue of 3xTg-PTS compared to WT-PTS as well as no significant difference in cleaved GSDMD expression between WT-PTS mice and 3xTg-Sham mice (Figure 6E). These results indicate that stroke in aged WT mice have a similar expression of inflammasome signaling as seen in AD mice without a stroke.

Figure 6. PTS induces inflammasome activation and GSDMD cleavage at 24 hours post-PTS in 3xTG and WT mice and 3xTg sham mice in intestinal tissue.

A) Representative images of western blot analysis showing the inflammasome protein distribution in the intestinal tissue of mice. Western blot analysis showing significantly increased cleaved caspase-1 (B), mature IL-1β (C), cleaved GSDMD (D) in intestinal tissue 24 hours after PTS in aged 3xTg compared to WT-PTS and 3xTg sham mice. C) ASC significantly increased in intestinal tissue 24 hours after PTS in aged 3xTg compared 3xTg sham mice. (n= 5–6 per group), *p<0.05, **p<0.01, ***p<0.001, One-Way ANOVA.

Damage to gut morphology occurs after PTS in 3xTg and WT mice and in 3xTg sham mice

Gut complications that occur in both AD and stroke animal models include gut-barrier breakdown (Arya & Hu, 2018; Das et al., 2022). We have previously published data demonstrating that gut permeability and gut barrier morphology are aggravated in young mice after stroke (Kerr et al., 2022). Here, we show that in aged WT-PTS mice, gut anatomy is worsened as shown by the wave-like morphology and damage to the microvilli compared to WT-sham mice (Figures 7A and B) 1 month after stroke. In addition, aged 3xTg- sham operated mice showed worsened microvilli morphology compared to WT-PTS mice (Figure 7C). These findings indicate that aged AD mice without focal cortical infarction present with evidence for gut barrier dysfunction. Furthermore, when PTS was produced in 3xTg mice, there was a dramatic increase in the infiltration of inflammatory cells into the gut epithelium (Figure 7D) compared to WT-PTS mice as well as 3xTg and WT sham mice. In addition, these mice also showed evidence of worsened microvilli morphology compared to WT-PTS mice as well as 3xTg and WT sham mice as shown by a reduced number of microvilli as well as a loss of a finger-like morphology (Figure 7D). The gut epithelium of aged WT-sham mice showed healthy microvilli morphology, while aged WT-PTS presented more evidence of wave-like microvilli morphology (Figures 7A and B). Aged 3xTg sham mice show worsened morphology compared to aged WT sham mice and PTS-WT mice (Figures 7B and C). Lastly, we also performed a FITC dextran gut permeability assay to confirm the histological results showing morphological changes in aged stroke WT and 3xTg mice (Figure 7E). Results demonstrated that aged WT-PTS mice had significantly increased gut permeability compared to aged WT-Sham mice. Interestingly, permeability in aged WT-PTS mice was comparable to 3xTg-sham mice, supporting our histological analyses. In addition, after the induction of PTS, 3xTg-PTS mice presented significantly increased permeability compared to 3xTg-Sham and WT-PTS mice.

Figure 7. PTS alters microvilli morphology and epithelium morphology as well as permeability changes in intestines in 3xTg and WT mice.

H&E sections showing morphological changes in intestinal histology including disruption of wave-like morphology of microvilli (arrows) in aged 3xTg (C and D) and WT-PTS (A and B) mice and increased infiltration (*) of inflammatory cells in aged 3xTg-PTS mice compared to WT-PTS mice 1 month after stroke. Magnification: 60X. Scale Bar = 10 μm. E) Gut permeability to the high molecular weight FITC-dextran (MW: 4kDa) is significantly increased in aged 3xTg-PTS mice compared to aged WT PTS and 3xTg sham mice and in WT-PTS mice compared to WT sham mice 72 hours after PTS. (n= 5–6 per group), *p<0.05, **p<0.01, ***p<0.001, One-Way ANOVA.

PTS leads to an increase in Aβ accumulation in aged WT mice

It has recently been demonstrated that peripheral Aβ exists in the intestines of AD-transgenic mice (Jin et al., 2023). Thus, after confirming that aged WT-PTS mice suffer gut morphology changes like that observed in the 3xTg sham mice, and that 3xTg-PTS exhibit more severe damaged gut epithelial morphology, we performed immunohistochemical analysis to examine the presence of Aβ accumulation in the gut. Intestinal sections were taken 1 month after PTS and stained for EpCam1 (a gut epithelial cell marker) and Aβ as a marker for AD pathogenesis. We found that Aβ is present in aged 3xTg-Sham mice as well as 3xTg-PTS mice in gut epithelial cells (Figures 8B and D). Interestingly, there did not appear to be worsened gut epithelial cell morphology or an obvious difference in the amount of Aβ between the two groups. However, when examining the accumulation of Aβ in gut epithelial cells in aged WT mice 1 month after PTS, we found increased Aβ compared to aged WT sham mice (Figures 8A and C). Furthermore, Western Blot analysis demonstrates a significant in increase in Aβ42 in 3xTg-PTS mice compared to WT-PTS mice at 24 hours post injury (Figure 8 E and F). Interestingly, there was no significant difference in Aβ42 between 3xTg-PTS mice and 3xTg-Sham mice (Figure 8 E and F). These results suggest that stroke in aged WT mice induces the presence of gut Aβ, a pathological response that may contribute to the development of AD.

Figure 8. Aβ Accumulation is increased in aged WT mice 1 month after PTS.

A) Aβ (red) is increased in the intestinal epithelial cells (EpCam 1- green) in 3xTg-Sham (B) and 3xTg-PTS mice (D) compared to WT-Sham mice (A) and AD-sham mice (C). Images are taken at 60x using confocal microscopy. Scale Bar = 10 μm. Colocalization of Aβ and epithelial cells (arrows).

Adoptive Transfer of stool derived EVs from PTS mice into naïve mice induces inflammasome expression in the brain

Our previously published study demonstrated that adoptive transfer of serum derived EVs from young PTS mice intravenously injected into naïve mice induced inflammasome expression in the gut (Kerr et al., 2022). These results emphasized an important role for systemic EV signaling in disturbing the gut-brain axis after stroke. To investigate whether gut-derived EVs after stroke contribute to neuroinflammation and progression of AD pathology, we performed an adoptive transfer protocol in which we injected stool-derived EVs from aged PTS mice into aged naïve mice via the tail vein. To confirm that we isolated EVs from the stool samples, we analyzed the samples by NTA and found that EVs were in the correct size (100–200nm) range to be considered EVs (Figures 9A and B). To further confirm the presence of stool-derived EVs we also examined samples using EM and confirmed the classical round EV morphology/size (Figures 9C and D). Next, we performed western blot analysis on stool derived EVs to determine the levels of inflammasome and AD proteins. The mature form IL-1β was present in stool-derived EVs from all groups. We also found an increase in mature IL-1β and Aβ in EVs of both 3xTg-PTS and 3xTg-Sham mice compared to WT-PTS and WT-sham mice (Figure 9E). Additionally, we injected (IV) stool-derived EVs from 3xTg-PTS, WT-PTS, 3xTg-sham and WT-sham into WT naïve mice respectively. Our findings indicate that the injection of 3xTg-PTS stool-derived EVs into WT naïve mice significantly increased Aβ and cleaved caspase-1 expression in the cerebral cortex compared to EVs injected from WT-PTS mice (Figures 9F to H). Additionally, injection of stool-derived EVs of 3xTg-sham mice into WT-naïve mice also significantly increased Aβ and cleaved caspase-1 expression in the cerebral cortex. Taken together these data indicate that the gut-brain axis involving stool-derived EVs plays a role in inflammasome signaling in AD even in the absence of brain injury.

Figure 9. Increase of Aβ and inflammasome proteins in cortex after adoptive transfer of stool-derived EVs.

A-B) Nanoparticle tracking analysis (NTA) confirming isolation of stool-derived EVs in the correct size range (nm). C-D) Electron Microscopy images of stool-derived EVs from WT (C) and 3xTg (D) mice showing size of morphology of EVs. E) Western blot analysis showing increased expression of mature IL-1β and A β in stool-derived EVs of PTS and 3xTg sham mice compared to stool-derived EVs of WT PTS and WT sham mice. F) Adoptive transfer of stool-derived EVs from 3xTg PTS mice in 3xTg naïve mice significantly increases caspase-1 in cortical tissue compared to adoptive transfer of stool-derived EVs from WT PTS mice into WT naïve mice and 3xTg sham mice into 3xTg naïve mice 24 hours after adoptive transfer. G) Adoptive transfer of stool-derived EVs from 3xTg-PTS mice in 3xTg naïve significantly increased Aβ in cortical tissue 24 hours after adoptive transfer H) Representative images of Western Blot analysis of adoptive transfer of stool-derived EVs from 3xTg-PTS mice in 3xTg naïve. (n= 5–6 per group), *p<0.05, **p<0.01, ***p<0.001, One-Way ANOVA.

Discussion

Stroke has been defined as a known risk factor for AD (Zhou et al., 2015). Factors that have been studied in the relationship between AD and stroke include genetic predisposition, vascular risk factors, and aging (Lambert & Amouyel, 2011; Vijayan & Reddy, 2016). Approximately 10% of patients that suffer from a first stroke develop some form of dementia and up to 60% of stroke survivors suffer from cognitive decline (Lo et al., 2022; Subic et al., 2017). Several studies have determined clinical relationships between AD and stroke and have recognized that they commonly present in the aging population (Zhou et al., 2015). However, the understanding of AD as a risk factor for stroke as well as the development of AD pathology after stroke is less understood. In this study, our findings show that PTS in aged 3xTg mice produce well demarcated cortical infarcts at 1 month slightly larger than WT-PTS mice as well as hippocampal involvement, which is not observed in WT-PTS mice. In terms of behavioral deficits, 3xTg-PTS and WT-PTS mice showed cognitive deficits compared with wild type mice. However, 3xTg-PTS mice had worsened motor outcomes compared to WT-PTS mice at early time points after injury as well more signs of freezing and anxiety, which are common symptoms in both early and later stages of AD (Bernard et al., 2024). We also report that 3xTgr-Sham mice demonstrated elevated levels of abnormal inflammasome activation and pyroptosis in the brain and gut compared to aged WT-Sham mice. The examination of the gut after PTS in the 3xTg-AD transgenic model also showed increased levels of inflammasome and AD associated proteins as well as increased gut permeability. The adoptive transfer of stool-derived EVs from PTS mice into WT mice induced significant increases in the levels of active cleaved/mature forms of inflammasome proteins, GSDMD, and Aβ in the cerebral cortices of recipient mice. Together these studies demonstrate that stool-derived EVs play a critical role in the inflammasome signaling that contributes to gut-brain axis disruption after stroke and AD.

It is well known that stroke, especially in the aged population, leads to increased risk of cognitive decline and reduced motor function (Lo et al., 2022; Sun et al., 2014). In this study, sensorimotor function test using the rotarod demonstrated there was reduced motor function in 3xTg-mice compared to other groups at acute time points after injury, but not at chronic time points. NOR object results did not show any significant differences in spatial memory between the PTS groups, therefore suggesting at spatial memory is not affected between WT and 3xTg mice after stroke. However, Open Field results demonstrated increased anxiety-like behavior in 3xTg-PTS mice compared to WT-PTS mice. Open Field testing was performed at 10 days post PTS and mice showed no motor differences at this time point. Therefore. we do not believe that motor function contributes to Open Field results. These findings support the conclusion that increased cortical atrophy in aged 3xTg-PTS mice may contribute to differences in motor cognitive impairments compared WT-PTS mice. In this study we used both male and female mice but did not assess possible sex-dependent differences in these various outcome measures. One of the reasons for this limitation is due to increased mortality in female mice compared to male mice after stroke, therefore making statistics for sex differences challenging. Because sex is an important independent variable in the pathogenesis of ischemic stroke, future studies will have to be conducted to consider this factor on the consequences of cortical infarction on the disruption of the gut-brain axis in AD transgenic mice.

The importance of neuroinflammatory mechanisms in both the pathophysiology of acute stroke and AD indicates that specific inflammatory processes may impact the incidence and severity of these two distinct but interrelated neurological disorders. For example, inflammasome-mediated pyroptosis has been implicated in the pathogenesis of both stroke and AD (Alishahi et al., 2019; Liang et al., 2022; Puleo et al., 2022). We have previously reported that inflammasome protein expression is increased after PTS in the cerebral cortex as well as in serum and serum-derived EVs of stroke patients (N. Kerr et al., 2018; Kerr et al., 2022). Additionally, we recently demonstrated that traumatic brain injury induces inflammasome activation and microglial dysfunction in the 3xTg mouse model compared to WT mice (Johnson et al., 2023). Western blot analysis showed that inflammasome signaling and pyroptosis are also increased in the brains of aged 3xTg-stroke mice compared to aged WT mice. Our results also revealed that inflammasome proteins and GSDMD are increased in WT-stroke mice compared to AD-sham mice. It is known that inflammasome activation is increased acutely after stroke (Puleo et al., 2022). Our present findings in aged WT-PTS mice indicate that stroke leads to a more robust neuroinflammatory response compared to 3xTg mice presenting AD pathology without stroke. It is well known that inflammasome signaling increases in aging (Sebastian-Valverde & Pasinetti, 2020). Here, we present novel data demonstrating that stroke in the aging population increases inflammasome signaling and the presence of AD pathology.

It is known that Aβ protein metabolism is involved in stroke pathophysiology (Goulay et al., 2020). Approximately two-thirds of patients suffer from post-stroke cognitive impairment and related clinical studies have revealed the presence of Aβ deposits in the autopsy brains of stroke patients (Melkas et al., 2014; Ouyang et al., 2021). A previous study has also shown increased levels of Aβ in other organs besides the brain in mice with a genetic predisposition towards AD such as in the heart (Cyr et al., 2024). In this study, 3xTg mice presented increased levels of Aβ compared to the other groups in the brain. Interestingly, we also found that aged WT-mice that underwent PTS developed brain Aβ comparable to 3xTg sham mice. These data suggest that neuroinflammation is exacerbated after PTS in aged AD and WT mice and that the response in WT-PTS mice is comparable to the neuroinflammatory response that is seen in AD without stroke. Taken together, these findings support the notion that Aβ accumulation after stroke is involved in the pathogenesis of AD in aging.

Recently several studies have demonstrated that the inflammasome is involved in gut-brain axis communication in neurological conditions such as stroke, AD, Multiple Sclerosis, and Parkinson’s disease (Buga et al., 2023; Kerr et al., 2022; Rutsch et al., 2020; Shukla et al., 2021). The intestinal epithelial barrier is critical in maintaining homeostasis throughout the entire body’s internal environment (Rodriguez-Colman et al., 2017). Evidence suggests that stroke disrupts intestinal epithelial barrier integrity which can lead to microvilli and gut epithelium damage (Stanley et al., 2016). Preclinical studies have demonstrated increased gut inflammation and worsened gut microbiota in AD transgenic mice (Brandscheid et al., 2017; Kowalski & Mulak, 2019; Sohrabi et al., 2021). Both stroke and AD patients often suffer from GI disorders such as inflammatory bowel disease (IBD), irritable bowel syndrome (IBS), and gut dysbiosis (leaky gut) (Arya & Hu, 2018; Kowalski & Mulak, 2019; Li et al., 2017; Nakase et al., 2022; Wang et al., 2001). Several studies have reported that the inflammasome is involved in gut microbiota-related inflammatory changes in AD (Rutsch et al., 2020; Shukla et al., 2021). In addition, our group recently reported that inflammasome-mediated pyroptosis plays a pivotal role in disruption of the gut-brain axis after stroke in young mice (Kerr et al., 2022). In the current study, aged WT-PTS and 3xTg-sham mice demonstrated similar evidence of damaged morphology including damaged microvilli and immune cell infiltration in the gut epithelium. 3xTg-sham mice and WT-PTS also had similar gut permeability measurements as well as expression of inflammasome proteins and GSDMD. These results indicate that inflammatory responses and functional changes in the gut of aged mice after PTS are comparable to those seen in non-injured aged 3xTg mice. Our mechanistic findings also show that gut derived EVs released after PTS contain inflammasome and AD proteins which can then be endocytosed by brain cells including microglia, leading to pyroptosis, progression of AD pathology and worsened neurological outcomes (Graphical Abstract).

Previous studies have suggested that changes in the gut microbiome can influence the production and aggregation of gut amyloid proteins (Friedland, 2015; Pistollato et al., 2016). For example, Jin and colleagues (2023) reported that the gut is a source of Aβ and that changes in the gut microbiota may upregulate Aβ production and contribute to AD pathology in a 5xFAD mouse model (Jin et al., 2023). Our studies confirmed a robust presence of Aβ in the gut epithelium of aged 3xTg mice, but also provide evidence of Aβ in the gut epithelium of aged WT mice. These findings indicate that age alone is an important variable that can result in the presence of gut Aβ deposits. We also found that PTS in aged WT mice produced an additional increase of Aβ accumulation within the gut epithelium like 3xTg sham mice. The 3xTg mouse model controls the genetic component of the development of AD pathology. The results generated in this study are impactful in emphasizing an important role of genetics in AD pathology. Our findings also provide strong evidence that aging as well as the presence of a stroke may also lead to worsened AD pathology. These results provide new information regarding aging alone as a significant variable in the pathogenesis of gut dysfunction.

An important aspect to consider when studying the gut-brain axis in neurological disorders is the bidirectional nature of this important relationship. Here, we present for the first time an EV-mediated bidirectional gut-brain axis and demonstrate how disruption of this axis under specific experimental conditions leads to increased pyroptosis associated with gut and brain dysfunction. We have previously shown that serum derived EVs play a role in the transport mechanisms of inflammasome proteins between the brain and distal organs (lungs, heart and gut) after TBI and stroke (Cyr et al., 2024; N. A. Kerr et al., 2018; Kerr et al., 2022). This mechanism supports the “top-down” portion of the bidirectional gut-brain axis (Arya & Hu, 2018), where EVs released from the brain affect systemic organ systems. As demonstrated in the Graphical Abstract, we expand on our previous work and provide new data supporting the “bottom-up” approach of the bidirectional axis, where gut-derived inflammatory mediators may alter the response of the injured brain structure and function.

There are limited studies that demonstrate that Aβ is present in stool. It was recently reported that Aβ is expressed in feces of AD transgenic mice (Jin et al., 2023). Here, we provide additional findings that stool contains EVs carrying a cargo of inflammasome proteins in aged 3xTg and WT mice as well as Aβ. Furthermore, the adoptive transfer of stool-derived EVs from 3xTg-PTS mice to a naïve 3xTg mice were taken up in the brain and increased inflammasome protein expression and Aβ in cortical tissue. To the best of our knowledge, these data represent the first evidence that after stroke, Aβ and inflammasome proteins are incorporated into stool-derived EVs for systemic release. This mechanism supports our hypothesis that the focal ischemic insult resulting in blood-brain barrier breakdown leads to release of EVs rs from damaged brain cells into the systemic circulation (blood). These EVs the gut and then may then be taken up by endocytic into gut epithelial cells. In turn stool-derived EVs released after stroke are taken up into the brain, leading to a future increase in inflammasome protein expression and Aβ accumulation (Graphical Abstract). Further studies are required to examine cell specific uptake of stool-derived EVs into brain cells including neurons and glial cells. Studies focusing on targeting stool-derived EV uptake into the brain are also necessary to examine treatment strategies targeting disruption of the gut-brain axis after stroke and AD (Kerr et al., 2022). Although we focus this study on the role of stool-derived EVs in bidirectional inflammasome activation in stroke and AD, we hope to perform further studies that may focus on microglial or neuronal-derived EVs. This will allow us to potentially distinguish between the initial neuroinflammatory response and the gut inflammatory response in terms of EV signaling.

Several mechanisms are involved in gut-brain axis communication. Neural pathways that are classically involved in this gut-brain axis include sympathetic and parasympathetic activation, the hypothalamic-pituitary-adrenal (HPA) axis, and activation of the innate immune response (Stanley et al., 2018). However, how the innate immune response contributes to worsened outcomes in the brain-gut axis after stroke in AD remains poorly understood. In this study, we focused on inflammatory mechanisms that may participate in the dysfunction of the brain-gut axis. In order to have a more comprehensive standing, it will be necessary to study changes in the microbiota and how this also contributes to gut-brain axis disruption after stroke in AD. Gut-microbiota-derived EVs isolated from Gram+ and Gram- bacteria have been studied as signaling molecules in CNS disorders, such as Parkinson’s disease (Sun et al., 2023). The study of stool-derived EVs and inflammasome signaling in cell-to-cell communication in gut disorders is a relatively new area of research. Although we have examined stool derived EV’s and effects on brain and gut inflammasome proteins, it will be important in the future to also evaluate bacterial content in stool-derived EVs to determine whether changes in microbiota in these EVs are contributing to stroke and AD pathogenesis.

In summary, we report that (a) cortical infarct size and location are similar in 3xTg compared to WT mice after PTS and that cognitive dysfunction is worse in both aged WT-PTS and aged 3xTg-PTS mice compare to sham mice, (b) PTS induces significant morphological, functional changes and pyroptosis in the gut of WT-PTS and 3xTg-PTS, (c) microglia and gut epithelial cells express AD-related proteins in both WT-PTS and 3xTg-PTS mice, and (d) adoptive transfer studies showed that stool-derived EVs play a role in the progression of neuroinflammation and Aβ pathology after stroke. Taken together these findings support the concept of an EV-mediated bidirectional gut-brain axis that contributes to the progression of AD pathology in the brain and gut after stroke. Finally, a key finding of this study is the pathological and functional damage in the brain and gut documented in aged WT are comparable to AD transgenic sham mice that did not suffer a stroke under the current experimental conditions. These data also suggest a dysfunctional bidirectional gut-brain axis as an important therapeutic target for the treatment of AD related disorders in stroke patients. Studies have demonstrated that targeting the gut microbiota through various probiotics and short chain fatty acids has improved stroke recovery in mice (Benakis et al., 2020; Chen et al., 2019). Our current results therefore provide further information that targeting the inflammasome-mediated EV signaling and pyroptotic cell death in the gut may also be beneficial in the improvement of stroke outcomes and contribute to mitigating the progression of AD pathology in the aging population.

Highlights:

• Gut-brain axis dysfunction can lead to worsened neurological function after stroke and in Alzheimer’s Disease (AD)

• Inflammasome signaling and pyroptosis contribute to gut-brain axis dysfunction after stroke in AD.

• Amyloid-beta pathology increases with aging after stroke without genetic predisposition to AD.

• Stool-derived Extracellular Vesicles (EVs) serve as a signaling mechanisms for gut-brain axis dysfunction after stroke and AD.